I have a 2x4 fully repeated measures design in which participants had to correctly state whether a target was absent or present in a visual field task, with different numbers of distractors.

the first independent variable is target presence/absence

the second independent variable is the number of distractors in the task - either 2,4,6 or 8. This is represented in the data set with present2, present3, present6, present8, absent2, absent4, absent6, absent8 as the column headers.

there are significant main effects of both target absence/presence and number of distractors. there is also an interaction effect. However my problem is how to break down this interaction effect. I know I can't use tests such as Tukey as this is a fully repeated measures design. I am also aware that potentially multiple t-tests can be used and a bonferroni correction can be made. However I'm not sure how many t-tests I would have to run and on what?

alternatively in my reading I have found a book which uses syntax to break down simple effects using manova for a repeated measures anova. when I do this, all the effects are significant (i.e. there is a significant effect of target absence/presence at each number of distractors).

(but would a bonferroni correction need to be applied to this?)

with reference to a plot of the data it looks like target absence impairs reaction times for all numbers of distractors but that this impact increases as the number of distractors increase. But if the impact of target absence is significant in the same way, (i.e. reducing reaction times) for each number of distractors, what does this mean in terms of an interaction?

Sorry if this isn't clear, I would really appreciate any help as I'm really confused as to how best study this interaction effect. any comments or suggestions would be very much appreciated.